search

pytorch / pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration
https://pytorch.org
Other
82.33k stars 22.14k forks source link

DISABLED test_attention_vs_linear (__main__.MemoryBudgetTest) #128217

Open jeffdaily opened 3 months ago

jeffdaily commented 3 months ago

Platforms: rocm

This test was disabled because it is failing on main branch (recent examples).

Caused by https://github.com/pytorch/pytorch/pull/126320. Seeing if disabling this one issue can bring HUD back to green. If not, a revert will be issued.

cc @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @dllehr-amd @jataylo @hongxiayang @ezyang @bdhirsh @anijain2305 @chauhang

pytorch-bot[bot] commented 3 months ago
Hello there! From the DISABLED prefix in this issue title, it looks like you are attempting to disable a test in PyTorch CI. The information I have parsed is below: * Test name: `test_attention_vs_linear (__main__.MemoryBudgetTest)` * Platforms for which to skip the test: rocm * Disabled by `jeffdaily` Within ~15 minutes, `test_attention_vs_linear (__main__.MemoryBudgetTest)` will be disabled in PyTorch CI for these platforms: rocm. Please verify that your test name looks correct, e.g., `test_cuda_assert_async (__main__.TestCuda)`. To modify the platforms list, please include a line in the issue body, like below. The default action will disable the test for all platforms if no platforms list is specified. ``` Platforms: case-insensitive, list, of, platforms ``` We currently support the following platforms: asan, dynamo, inductor, linux, mac, macos, rocm, slow, win, windows.
bdhirsh commented 3 months ago

hmm @Chillee looks like after the memory budget changes in the partitioner, those tests started failing under rocm 🤔